Impacts of climate modes on temperature extremes over Bangladesh using statistical methods

Bangladesh is sensitive to weather and climate extremes, which have a serious impact on agriculture, ecosystem, and livelihood. However, there is no systematic investigation to explore the effect of climate modes on temperature extremes over Bangladesh. A total of 11 temperature extreme indices based on the daily maximum and minimum temperature data for 38 years (1980–2017) have been calculated. Cross-wavelet transform and Pearson correlation coefficient have been used to identify the relationship between temperature extremes and three climate modes namely El Niño Southern Oscillation (ENSO), Indian Ocean Dipole (IOD), and North Atlantic Oscillation (NAO). Detrended fluctuation analysis (DFA) method was applied to predict the long-term relationship among temperature indices. Results showed that warm (cold) temperature extreme indices increased (decreased) significantly. There was a significant upward trend in the diurnal temperature range and tropical nights except for growing season length. ENSO and IOD had a strong negative impact on warm temperature indices, whereas NAO had a strong negative influence on variability temperature indices in Bangladesh. Temperature extreme had a long-term relationship based on DFA (a > 0.5), implying that the temperature extremes will remain their present trend line in the future period. The Poisson regression model showed that the highest probability (65%) of having a 2–4 warm spell duration indicator (days/decade) is consistent with the observation, which is shown in the cross-wavelet transform and spatial analysis.